First LEON Silicon Tested Successfully

uglomera writes: "LEON, the open-source CPU developed for the European Space Agency, has been successfully manufactured and tested by Atmel on an Atmel ATC35. Gaisler Research, whose CEO Jiri Gaisler wrote the VHDL model of LEON, also offers a real time kernel, simulator, a cross-compiler, etc. for this SPARC-family processor designed for space applications. Check it out." You can find more good information on the LEON processor on the Gaisler site, including diagrams and further reading. Open Source hardware running Free software -- wheee!

Should we trust space flights to open source?

I have a few minor reservations about using an open source developed processor for missions in space. Open source is great, and produces some of the most well written, secure code that is available, anywhere, but I'm not sure we should be trusting this technology to aide in piloting a space vehicle. A space flight is a dangerous and EXTREMELY complicated process, and if a single component fails, or produces an error, the results can be catastrophic.

NASA, in the USia, spends literally millions of dollars designing the space shuttle's computing system from the ground up. Their in-house coders pour over each microchip and line of code hundereds of times looking for even the smallest bug. This superior attention to detail is not possible using the limited resources of the open-source method. It is also why NASA can succeed in safe, reliable space flight time, and time again, while other space programs are struggling. Open source has it's advantages, but when 100% reliability is necessary, it may not be the best option.

OpenCores

More stuff like this can be found here: opencores [opencores.org]

they will test it, same as a commercial processor

I doubt they plan to send this widget up without a full seperate functionality audit, which they can do for an open source processor, but it might be hard to convince say Intel to let them pore over the VHDL for the celeron or whatever.
Just because it's open source doesn't mean they aren't going to put it through the same rigorous tests that they would put a commercial processor through.

Open CPUs vs Other Open Hardware

I was speaking to a co-worker the other day about Sun's UltraSPARC III processor and he was telling me about CPU manufacturing in general.

To actually create a fabrication facility to make CPUs, it takes about 20 billion dollars. $20,000,000,000 dollars. That's more than most companies can afford. Even Intel couldn't make very many new fabrication plants.

If a company can't afford to create their own plant, they have to schedule time at a fabrication facility. This is basically a window (say 48 or 72 hours) where the facility will crank out as many chips as possible. If they miss the window for some reason, they have to re-schedule and it can be months until there is another open time slot.

What I'm getting at is designing Open CPUs is a great idea. It allows developers to really get inside of the hardware and optimize the hell out of applications, which is a good thing. However, the actual cost to make these CPUs is staggering, and unless a big company put up some big bucks, I don't see it happening in the near future.

My co-worker also mentioned how low cost almost everything else is. Video cards, NICs, sound cards and the like. Wouldn't it be better to focus on products like these since they would work with all hardware (how about a video card that worked on Sun and x86 machines?)? With Moore's Law getting us faster and faster CPU speeds, perhaps it's time to make the peripherals first, and focus on a CPU once we have found sucess with smaller projects.

Fault tolerant for hardened environments

Please to read the article(s) before firing off
negative spin.

LEON [gaisler.com]

The LEON model also exists in a fault-tolerant
version, capable of detecting and correcting
single-event upset (SEU) errors in any register
or RAM element. This is done completely in
hardware without any software intervention.
The area overhead for the fault-tolerance
functions is approximately 30% while the timing
penalty is around 5%. The fault-tolerant features
makes it possible to use LEON in the severe
space environment without having to develop specific SEU-hardened cell libraries. The LEON
fault-tolerant VHDL model can under some conditions be licensed from ESA -

Re:"Open Source" hardware

> I guess my feelings are known because I believe that hardware's design should not be completely open

Would you care to explain us why ?

I see advantages in the fact that:

1/ Everyone (with the knowledge to understand it) can see how the processor is done
2/ Everyone (with the motivation) can learn from the processor core
3/ Everyone (with the foundry) can manufacture a totally compatible processor
4/ Everyone (with the technical skils) can improve the procesor design
5/ (A few other ones)

Of course you are not going to run a plant in your backyard, but I fail to see real downsides to the idea.

Cheers,

--fred

Re:Open CPUs vs Other Open Hardware

If you want highly cost efficient production in huge quantities, you are right. If you want just one, you can use an FPGA (the article specifically mentions using Xilinx chips) and reprogram it at will - great for testing and development, or if you want a small series there are a number of companies producing small series of chips.

The downside? Cost and performance. You won't get high-end performance out of a FPGA, for instance. But on the other hand, for an embedded system an FPGA give you the opportunity of "easily" developing a full system on a chip, test it, debug it in actual hardware, and depending on the volume you need decide whether to use FPGAs or custom ASICs in the final, shipping device.

Important for Spaceflight Applications

The availability of an inexpensive, radiation tolerant CPU is a big win for space researchers. Right now there are darn few radiation-tolerant parts available for use in space applications due to decreased demand from the military. The International Space Station is using Intel 386s for embedded CPUs, as they are simple enough to be relatively rad-hard. More modern CPUs, such as in the laptops used on the ISS and Shuttle have about one lockup/day due to radiation.

The design requirements for software controlled systems in space are so stringent that to do anything sophisticated requires incredible redundancy, cross checks among the systems, and increased design complexity, all of which significantly drives up design costs (and causes all kinds of debugging problems). Tell me three times is not enough, you have to tell three controllers three times, three different ways and then they need to cross check. This could be a big step forward for software geeks in space.

Could you check that number?

$20 billion? Hell, an aircraft carrier only costs about $3 billion. IIRC the newest Intel fab cost about $2 billion.

Ummmm....what?

"Their in-house coders pour over each microchip and line of code hundereds of times looking for even the smallest bug. This superior attention to detail is not possible using the limited resources of the open-source method."

Sure it is, here's how: Use exactly the process NASA uses know and that you are apparently comfortable with. Then ADD (not replace) more programmers by making the source available via FTP.

Adding openness to an existing project loses nothing. Yes, shifting the burden of quality OFF of some process ONTO openness may not always be a good idea (not in one go, anyway). But adding more checks doesn't lower quality.
--

Re:Should we trust space flights to open source?

"A space flight is a dangerous and EXTREMELY complicated process, and if a single component fails, or produces an error, the results can be catastrophic. "

I don't know the rules that the ESA has to live with, but if they're anything like the rules the DOD imposes on rocket launches, if something goes wrong, you just blow up the rocket. Even the shuttle's SRBs have the equivalent of a really long stick of dynamite to make sure that, in case of an accident, no pieces bigger than my hand or so would ever reach the surface.

"NASA, in the USia (sic), spends literally millions of dollars designing the space shuttle's computing system from the ground up. "

We're talking about the ESA here, not NASA. There are only two countries out there that have manned spaceflight programs, and PRC is much closer to being number 3 than any European nation or group. In my opinion, the Japanese will have manned space flight before the Europeans.

Today's useless fact: After the US and Russia/CIS, the country with the most people that have gone into space is Canada.

If you're going to compare the uses of these chips to an American launch platfrom, I'd use the Titan III, or maybe the Delta 2, but definately not the STS.

"Their in-house coders pour over each microchip and line of code hundereds of times looking for even the smallest bug. This superior attention to detail is not possible using the limited resources of the open-source method. "

Tell that to the NSA. So far, it seems they're doing pretty well at scrutinizing Open Source operating systems (ie. GNU/Linux), and seem to be on the verge of making it the most secure modern operating system hands-down.

Open Source means that the hacker sitting in his mom's basement with a computer and a bag of Fritos can (legally) take apart and tinker with the innards. It doesn't REQUIRE you to fit that stereotype, though. If it did, then what does that say about IBM's efforts to run Open Source software on their mainframes?

"It is also why NASA can succeed in safe, reliable space flight time, and time again, while other space programs are struggling."

I think you're confusing "attention to detail" with "multi-billion dollar infrastructure set up in the paranoid 50's and 60's to support the Saturn V." As long as they had the money to pay for the rockets, you could give Brevard County, Florida to any country in the world and they'd be able to build and launch super-heavy lifters to their heart's content.

There's so much there that even Florida's Spaceport Authority (part of the Florida Department of Transportation) owns and maintains their own launch facilities on the Cape (Launch Complexes 20 and 46, I believe). http://www.spaceportflorida.com [spaceportflorida.com]

Getting into space doesn't require millions of dollars. Currently, all it requires is hundreds of dollars in model rocketry parts (and permission from the DOD and FAA to launch it). Sure, getting into orbit is even trickier, but I know of a team at Embry-Riddle Aeronautical University in Daytona Beach working to do just that with a 6' rocket. The only expensive part of the equation is trying to put something the size of a Mack truck into orbit.

... and that concludes my rant.

Re:Should we trust space flights to open source?

On the other hand, NASA can't tell the difference between feet and meters (they've botched that one up more than a few times), rely on antique systems because hardware manufacturers can't divide, and haven't the money to build a next-generation of vehicles (X-33, X34, the blended-wing aircraft, to name but 3 projects axed).

The ESA is not much better, having blown up at least one Ariane rocket because they'd swapped a + for a -. But at least, that was unmanned and merely cost a lot of people a lot of money, rather than their lives.

Open Sourcing means more eyeballs and more checking. Checking that space agencies (running on VERY VERY tight deadlines) can't afford to do. Sure, the patches will need to be examined ruthlessly, but with exhastive testing possible via simulations on every PC and its cousin, deliberate or accidental bugs can be all but wiped out.

Don't trust an agency because it's big. Trust it because it's done something to merit it. The NSA, for example, has probably done more for it's image in the past few months, via SE Linux, than it has ever achieved in the rest of it's existance. The ESA's move could, likewise, turn this barely-known, insignificant launcher of commercial satellites and the odd scientific one, into a major space research organization. COULD. Not will. There's a long way to go, yet, but getting known is a good step in the right direction. Getting known and aided by Nerdius Technomaximus is even better.

P.S. Anyone want to bet on when Linux'll get a self-modifying architecture?